2015 Informs Annual Meeting

MC56

INFORMS Philadelphia – 2015

MC55 55-Room 108B, CC Efficiency in the Higher Education and Banking Sectors Cluster: Data Envelopment Analysis Invited Session Chair: Jill Johnes, Professor, University of Huddersfield , The Business School Queensgate, Huddersfield, HD1 3HG United Kingdom j.johnes@hud.ac.uk 1 - Does Bank Performance and Corruption Matter for Economic Growth? An International Exploratory Study Abdel Latef Anouze, Assistant Prof, Qatar University, College of Business and Economic-DMM, Doha, No, 2713, Qatar, a.anouz@qu.edu.qa While previous economic development studies take into account the impact of financial sector performance on economic growth, the role of corruption as economic factors remain yet to be assessed. This paper integrates DEA and Structural Equation Modelling (SEM) to explore these relationships. Experimental experiences are reported on a sample of banks from different countries. The analytical results provide illustration on how to integrate DEA and SEM to examine and predict economic development 2 - Bank Branch Operational Performance: A Robust Multivariate and Clustering Approach Kostas Triantis, John Lawrence Professor, Virginia Tech, 7054 Haycock Rd, Room 428, Falls Church, VA, 22043, United States of America, triantis@vt.edu, Joseph Paradi, Haiyan Zhu, Oscar Herrera-restrepo, William Seaver We investigate bank branch operational performance by integrating robust techniques for clustering analysis and data envelopment analysis (DEA). By applying robust techniques based on principal component analysis, we look for the detection of branches exhibiting extreme operating behaviors (i.e., influential), and the clustering of branches based on operating characteristics. Our premise is that influential branches affect both the clustering and the determination of efficiency performance. 3 - Costs and Efficiency in the English Higher Education Sector: Latent Class Stochastic Frontier Models Jill Johnes, Professor, University of Huddersfield , The Business School Queensgate, Huddersfield, HD1 3HG United Kingdom j.johnes@hud.ac.uk, Geraint Johnes Using panel data from 2002/03 to 2010/11 we estimate a cost function for English higher education institutions (HEIs) using the latent class variant of the stochastic frontier model. We are able to: identify clusters of institutions based on the data; evaluate the parameters of the cost function for each cluster; estimate economies of scope and of scale within each cluster; measure efficiency of each institution relative to all HEIs and other HEIs in the same cluster.

3 - Request for Service Process (RFS) Process Reengineering for IBM Strategic Outsourcing (SO) Business Pawan Chowdhary, Software Research, IBM Research, 650 Harry Road, San Jose, CA, 95120, United States of America, chowdhar@us.ibm.com, Jeanette Blomberg, Anca Chandra RFS’s are small work items that were not covered in strategic SO contract and are highly profitable. For a typical contract there could be hundreds of RFS that needs to be executed. The traditional processes to manage these RFS’s were very cumbersome and often got delayed. In this talk, I will present the findings and the process reengineering work that IBM Research did to improve the performance of the RFS’s along with increased customer satisfaction. MC53 53-Room 107B, CC Opportunities and New Directions for Behavioral OM Sponsor: Behavioral Operations Management Sponsored Session Chair: Stephen Leider, University of Michigan, 701 Tappan Ave R4486, Ann Arbor, MI, 48104, United States of America, leider@umich.edu 1 - Supply Chain Risk: Behavioral Research Opportunities Brian Tomlin, Professor, Tuck School of Business, Dartmouth College, 100 Tuck Hall; Hanover, NH, United States of America, Brian.T.Tomlin@tuck.dartmouth.edu I will give an overview of the theoretical research in supply chain risk and attempt to identify some interesting questions for behavioral research. 2 - Humans Versus Machines: Impact on Network Capacity Jan Van Mieghem, Professor, Kellogg School of Management, 2001 Sheridan Road, 5th Floor, Evanston, IL, 60201, United States of America, vanmieghem@kellogg.northwestern.edu, Itai Gurvich, Lu Wang One of the fundamental questions in operations is to determine the maximal throughput or productivity of a process. Does it matter whether humans or machines execute the various steps in the process? If so, how do we incorporate this difference in our planning and performance evaluation? We propose some answers by discussing two examples: a theoretical analysis an empirical study. 3 - Sustainability: Challenges and Research Opportunities Erica Plambeck, Professor, Stanford Graduate School of Business, 655 Knight Way, Stanford, CA, United States of America, elp@stanford.edu Environmental sustainability requires profound changes in the production of goods and services. Behavioral OM researchers can guide and learn from those changes.

MC54 54-Room 108A, CC Equilibrium Routing and its Paradoxes

MC56 56-Room 109A, CC Location Analytics Sponsor: Location Analysis Sponsored Session

Cluster: Tutorials Invited Session

Chair: Asu Ozdaglar, Massachusetts Institute of Technology, 32 Vassar St, Cambridge, MA, United States of America, asuman@mit.edu 1 - Tutorial: Equilibrium Routing and its Paradoxes Asu Ozdaglar, Massachusetts Institute of Technology, 32 Vassar St, Cambridge, MA, United States of America, asuman@mit.edu We study equilibrium routing of flows in congested traffic and communication networks. We investigate efficiency implications of decentralized routing with and without prices and with different information structures. Despite the tractable nature of the models, both the equilibrium properties and the potential types of inefficiencies are rich and can sometimes change in unexpected directions in response to modifications in network and information structures.

Chair: Alan Murray, Professor, Drexel University, 3141 Chestnut Street, Philadelphia, PA, 19104, United States of America, amurray@drexel.edu 1 - Locating Units in a Data Network with Full Reliability and Redundancy

Sergio García Quiles, Lecturer In Operational Research, University of Edinburgh, James Clerk Maxwell Building, Peter Guthrie Tait Road, Edinburgh, EH9 3FD, United Kingdom, sergio.garcia-quiles@ed.ac.uk, Lukas Schaefer, Andreas Mitschke, Vassili Srithammavanh We study the problem of designing a data network that has to fulfil some restrictions while being optimal under certain criteria. Some data flows must be sent between certain units to be located and there must be full reliability and redundancy . A network is fully reliable if the probability of any given function failing is less than a given safety threshold. Full redundancy means that no single failure leads to the loss of any function. This problem is motivated by a real application.

223